AR-Tri-Training: Tri-Training with Assistant Strategy

Abstract:

The classification performance of the classifier is weakened because the noise samples are introduced for the use of unlabeled samples in Tri-training. In this paper a new Tri-training style algorithm named AR-Tri-training (Tri-training with assistant and rich strategy) is proposed. Firstly, the assistant learning strategy is posed. Then the supporting learner is designed by combining the assistant learning strategy with rich information strategy. The number of mislabeled samples produced in the iterations of three classifiers mutually labeling are reduced by use of the supporting learner, moreover the unlabeled samples and the misclassified samples of validation set can be fully used. The proposed algorithm is applied to voice recognition. The experimental results show that AR-Tri-training algorithm can compensate for the shortcomings of Tri-training algorithm, further improve the testing rate.

• Periodical: Applied Mechanics and Materials (Volumes 513-517)
• Pages: 1840-1844
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.513-517.1840
• Online since: February 2014
• Authors: Long Jie Cui*, Hong Li Wang, Rong Yi Cui
• Keywords: Assistant Learning Strategy, Rich Information Strategy, Semi-Supervised Learning, Tri-Training

References

[1] Kunlun Li, Wei Zhang and Yunna Dai: Computer Engineering and Applications. Vol. 45-22 (2009), pp.103-106 In Chinese.

[2] A. Blum, T. Mitchell: Combining labeled and unlabeled data with co-training. Proceedings of the 11th annual conference on Computational Learning Theory. Wisconsin, USA, 1998: 92-100.

DOI: 10.1145/279943.279962

[3] S. Goldman, Y Zhou: Enhancing supervised learning with unlabeled data. Proceedings of the 17th International Conference on Machine Learning. California, USA, 2000: 327-334.

[4] Zhihua Zhou, Ming Li: IEEE Transactions on Knowledge and Data Engineering. Vol. 17-11 (2005), pp.1529-1541.

[5] K. Nigam, A.K. McCallum, S. Thrun and T. Mitchell: Machine Learning. Vol. 39(2-3) (2000), pp.103-104.

[6] A. Blum, S. Chawla: Learning from labeled and unlabeled data using graph mincuts. Proceedings of the 18th International Conference on Machine Learning. Williamstown, MA, 2001: 19-26.

[7] Chao Deng, Maozu Guo: Chinese Journal of Computers. Vol. 30-8 (2007), pp.1213-1226 In Chinese.

[8] Hua Bi, Hongli Liang and Jue Wang: Chinese Journal of Computers. Vol. 32-5 (2009), pp.862-877 In Chinese.